DESCRIPTION: (Applicant's Description) Comparative genomic hybridization (CGH) has demonstrated its ability to provide a unique insight into the genetic abnormalities involved in the development of cancer. By co-hybridizing total genomic DNA from a specimen and from normal cells to metaphase chromosomes, it provides a genome-wide map of the DNA sequence copy number variation with cytogenetic scale resolution, - 20 Mb. Application of CGH to a wide variety of tumor types has found that most of them contain copy number changes that recurrently affect many regions of the genome, indicating the locations of known and many yet to be discovered oncogenes and tumor suppressor genes. We have recently demonstrated the feasibility of a new form of CGH that detects copy number variations comparative hybridization to a DNA microarray containing clones from any set of well mapped loci. Resolution with array CGH is more than 100 fold better than with standard CGH and the copy number variations are mapped relative to the genetic and physical maps being produced by the human genome project. We are proposing to develop a robust, flexible implementation of array CGH the human genome. This will require 1) selecting appropriate target clones with a goal of providing 1 Mb resolution (-3000 clones) throughout the genome. 2) improving methods for making arrays, and for acquiring and analyzing fluorescence signals. 3) improving methods for hybridization and for genomic DNA amplification to minimize the amount of specimen needed for analysis and to assure accurate results. 4) developing informatics for managing the project and displaying and analyzing the data. The project is designed so that it begins to provide useful measurement capability in its first year, and continues to improve throughout the 5 year project period. The performance of array CGH will be tested within the project as is necessary for technical development, and it will be evaluated in practice in collaboration with the national breast cancer SPORE projects. In summary, this project will develop the technology to permit human genome-wide, high throughput analysis of DNA copy number. The availability of this technique will facilitate positional cloning of cancer genes, and very large scale studies of the correlation of genetic abnormalities with biological and clinical behavior of tumors. The combination of array CGH data with data from other sources, for example microarray measurements of mRNA expression. will provide an even more powerful window on the genetic events involved in cancer development and progression.